Writing SQL queries on top of your database is not a novel idea, in fact I’m pretty sure it was one of the default sandbox experiences in OpenAI whenever they first opened access up to their API. We never saw that as our product either, it’s an expectation, a feature, and it’s why we open sourced our version over a year ago now. Don’t get me wrong, it is amazing, it supports anyone to query their database without being a SQL expert, and it enables those who know SQL to work even faster. That being said it was always a building block feature that we wanted to evolve, and now we have.
Introducing Outerbase North Star
(You also get a sneak peek at our all new charting library, coming soon 🤫)
Introducing Outerbase North Star. With North Star we are taking the next logical step after query generation, and enable AI to actually do analysis on top of your data. You simply need to define what matters most to you, is it revenue, sign ups, churn? You provide the metric and let our AI agent handle the rest. Outerbase will auto-generate dashboards, provide timely metrics, and even insights into what is potentially causing the changes. You can schedule the frequency of reports, how you want to be notified Slack, Email, or in app. Additionally you can improve the metrics by letting Outerbase know what type of charts and insights are helpful and further fine-tune the results that you receive.
It also becomes iteratively more helpful over time based on the type of data it is navigating and reporting on each week. You can see historical differences, trends and time-based comparisons. Allowing you to understand what went right each week or in some cases what didn’t. This will prove invaluable to users trying to recreate those sparks of success or digging into the moments that didn’t go as planned.
How it Works
Users who are interested in defining their North Star can opt-in to having their data analyzed (for our standard AI functionality all we use is your schema), and then define their metric. Outerbase AI can help you choose from potential metrics based on your data. From there sit back and let Outerbase do its magic, the initial reporting takes a little longer than the subsequent reports as it generates the functionality entirely and it takes into consideration your entire data set. From there it will generate an initial dashboard where you can pin what is helpful and dismiss what isn’t. Next you’ll set up your frequency of reports and Outerbase will help showcase the difference between each report and provide insights into what’s causing it. You can schedule these reports to be sent out via Email or Slack.
Security
Now your question might be “Why would I want any AI crawling through my data?”, and that is a completely valid concern. I do want to reiterate though that we never store your data in any of our servers, this is a completely opt-in feature, and that even when you do engage with it, we take a point in time query of the data, analyze it, and then only store the results of the insights. Meaning your data is still yours, we aren’t going to use it for any training purposes, and we definitely aren’t going to store it.
What’s Next
This is just the beginning of AI insights into your data, we want to further improve this with forecasting, more powerful chart components, collaborative experience with comments and callouts, and a bunch of other exciting features. As you can see this is just the groundwork of what’s to come with data insights and we’re so excited to bing you along for the journey.
Getting Access
North Star is still in private beta and only open up to select paying customers, we’re doing our own reporting and insight gathering before we open it up to the broader public. If this is something you’re interested in please reach out, tell me your use case and if it makes sense we can get you bumped to the top of the waitlist. Happy exploring!